The £20,000 to £30,000 question
Is it time to recalibrate NICE’s threshold for decision making?
Two decades have passed since the primary decision-making body in the UK for gaining access to new medicines, the National Institute for Health and Care Excellence (NICE), revealed what was widely suspected, that there was an explicit threshold being used to guide its decision making. No single number was given, instead a range was proposed that reflected two inflection points. Below £20,000 per QALY, it was argued that health technologies (not just pharmaceutical, but in reality, most new technologies NICE reviews are drugs) are likely to be considered cost-effective for use in the NHS. Above £30,000 per QALY, it would be unlikely that a new technology would gain approval.
No doubt acutely aware of pressure from across the pond for European countries to pay their ‘fair share’ for pharmaceutical innovation, the Association of British Pharmaceutical Industries (ABPI) has called for an urgent review of NICE’s cost-effectiveness threshold. They argue that it is inappropriate for this threshold to stay static for twenty years and argue that this stagnation is harming the UK’s competitiveness on the global stage.
But has the cost-effectiveness threshold really remained static for the past two decades? The apparent stability hides a quiet drift. In practice, the system has evolved a series of workarounds that has raised the effective threshold by stealth. These accommodations, including the Highly Specialised Technologies (HST) route for rare diseases, the Cancer Drugs Fund (CDF), and now the Innovative Medicines Fund (IMF), have allowed higher prices to enter the system without an explicit revision of the threshold itself. Appraisals within the standard process increasingly cluster toward the upper end of the range and now include a severity modifier (replacing the previous end-of-life criteria), implying tacit flexibility even without formal rule changes.
This strategy of accommodation avoids direct confrontation with the financial and ethical implications of changing the threshold. However, it produces serious distortions. First, it undermines transparency: patients, clinicians and the general public cannot see how value is determined. Second, it compromises equity: disease areas fortunate enough to attract bespoke mechanisms (most visibly cancer) gain access to treatments denied elsewhere (while the IMF was designed to alleviate this concern, doubts have been raised as to its effectiveness). Third, it erodes efficiency: ad hoc exceptions weaken the principle that equal health gains should receive equal priority. A more honest approach would acknowledge that the implicit threshold has already risen and formalise it transparently, replacing ad hoc exceptions with a single consistent rule.
The common objection to a higher threshold is fiscal. Critics fear it would drive up NHS drug spending and threaten financial sustainability. That concern is overstated. As long as NICE continues to apply rigorous cost-effectiveness analysis to determine which patients should be prioritised to receive new medicines, a modest upward adjustment in the threshold would relax restrictions slightly but not unleash uncontrolled spending. Moreover, the United Kingdom already operates a mechanism to contain the overall medicines budget. The Voluntary Scheme for Branded Medicines Pricing and Access (VPAS), successor to the Pharmaceutical Price Regulation Scheme, caps industry-wide expenditure through rebate payments based on aggregate spend, decoupling affordability from threshold setting. As long as that system functions effectively, the NHS can tolerate some flexibility at the margin without exceeding its total budget envelope. In this sense, the threshold determines value, while VPAS ensures affordability.
It is sometimes argued that a higher threshold could indirectly stimulate innovation by signalling greater willingness to pay for future therapies. Yet the empirical link between national pricing and global R&D investment is uncertain and likely weak compared with other determinants such as scientific capability, workforce, and regulatory stability. The stronger justification for recalibration lies not in rewarding innovation but in restoring transparency and coherence to NICE’s decision framework. The economic returns from a vibrant life sciences sector – including productivity gains from improved population health – should be addressed through dedicated industrial and fiscal policy instruments. The UK already promotes pharmaceutical R&D through targeted supply-side incentives – research funding, R&D tax credits, and patent protection. NICE’s role is to assess the health value of individual technologies at the point of decision, not to engineer macroeconomic incentives. Conflating the two weakens both policy domains.
If recalibration is accepted, the next question is how to update the threshold over time. The ABPI has argued for a one-off correction followed by indexation to inflation. While simple, that approach risks over-compensating during periods of high inflation and fails to reflect changes in the country’s real capacity to pay for health. An alternative is to link the threshold to per-capita GDP, which captures growth in national income and hence the resources society can allocate to health without displacing other priorities. The two indexing methods have diverged sharply over the past two decades. Inflation-indexing would effectively double the threshold to £40,000 to £60,000 while GDP-indexing would raise it by a more modest 40%, aligning it with improvements in living standards rather than with cost inflation. The latter provides a more sustainable and socially grounded anchor, which is also consistent with ABPIs argument that the the companies they represent contribute to UK economic growth.
Finally, recalibration should be accompanied by the withdrawal of special accommodations – such as HST, CDF & IMF, and severity modifiers – that fragment the system and obscure decision-making. A single, transparent threshold linked to per-capita GDP and reviewed every five years would restore coherence to NICE’s framework, maintain public trust, and reduce the administrative complexity of multiple exceptional routes.
NICE’s threshold has not stood still, it has drifted upward through a patchwork of exceptions that undermine fairness and clarity. An explicit, modestly higher threshold – indexed to GDP per capita and integrated with the existing VPAS expenditure cap – would be more transparent, equitable, and sustainable. After twenty years of quiet accommodation, the case for an open reassessment is compelling. NICE’s threshold should evolve deliberately, not by stealth.
A final note: The health economics community must not only help clarify what the threshold represents, but also ensure that its credibility is maintained over time. Without an active and critical professional debate, the threshold risks drifting again –quietly and without public scrutiny.





Thanks for the post. For sure, we need health economists to be more active in this debate. I also wrote about this myself today:
https://aheblog.com/2025/10/14/cost-effectiveness-thresholds-under-political-pressure-where-are-the-health-economists/
I would challenge your opposition to modifiers. Rejection of modifiers demands that we accept that a QALY is a QALY is a QALY, which is surely not always the case. Unless, of course, we expect value-based differential pricing below a maximum threshold, but that would mean modifiers by stealth. That said, I do think modifiers should be applied to outcomes (i.e. QALYs) rather than to decision thresholds, in which case the threshold could indeed be singular.
Thanks Andy and Francis, I’m trying to understand the impact of increasing the threshold, given the VPAS, which according to my interpretation of the google AI overview, restricts spending growth on branded medicines to 2% per annum across the drugs supplied by 172 companies (representing 85% of branded drug spending). I assume the 2% limit is reached every year? So increasing the threshold would mean prices go up and more new drugs are funded and used, but spending still only increases by 2%: companies pay more in rebates, so the NHS and patients are getting more and better drugs for the same cost as they would have incurred if the threshold had not been increased? Surely not, what am I missing?